Combination of the preoperative PSA level, biopsy gleason score, percentage of positive biopsies, and MRI T-stage to predict early PSA failure in men with clinically localized prostate cancer.

OBJECTIVES: Early (2 years or less) prostate-specific antigen (PSA) failure has been shown to predict for distant failure. The independent clinical predictors of time to postoperative PSA failure were used to identify prostate cancer patients at high risk for early PSA failure.
METHODS: A Cox regression multivariable analysis was used to determine whether additional predictive information was provided by the endorectal coil magnetic resonance imaging (erMRI) T-stage when controlling for the established prognostic factors in predicting the time to postoperative PSA failure in 977 men with palpable (T2) or PSA-detected (T1c) prostate cancer.
RESULTS: Preoperative PSA (P = 0.0001), percentage of positive prostate biopsies (P= 0.0001), erMRI T-stage (P = 0.0001), biopsy Gleason score (P = 0.0015), and clinical stage T2c disease (P = 0.004) were independent predictors of time to postoperative PSA failure. Two-year PSA failure rates derived from the Cox regression model and bootstrap estimates of the 95% confidence intervals are presented in nomogram format stratified by the preoperative PSA, percentage of positive prostate biopsies, erMRI T-stage, and the biopsy Gleason score.
CONCLUSIONS: Patients at high risk for early PSA failure and subsequent distant progression include men with erMRI T3 disease and 3 or more of 6 cores positive for a Gleason score 6 or higher disease when the PSA is more than 10 but not more than 20 ng/mL and any Gleason score when the PSA is more than 20 ng/mL. Men with erMRI T2 disease and 3 or more of 6 cores positive for a Gleason score 8 or higher disease and who have a PSA more than 20 ng/mL are also at high risk. Neoadjuvant therapy trials in these select patients may be justified.